SP 11.14 Anastomotic Leak Prediction after Colorectal Anastomosis (ALPACA) Study: Preliminary Results of a Prospective Multicentre Study

Abstract

Anastomotic leaks (ALs) after colorectal surgery pose major clinical challenges. Earlier detection could alter patient outcomes, q`az\ preventing the septic sequelae that follow ALs. Our aim is to identify biomarkers that positively predict ALs on days 1-2 post-operatively. Patients undergoing left-sided resectional surgery for benign or malignant pathology have been recruited since October 2024. Peritoneal drain fluid and blood were collected on post-operative days 1-2. Cell populations were characterised by multicolour flow cytometry and cytokine concentrations measured with the Siemens Healthineers IMMULITE® 2000 Immunoassay system. As of January 2026, 95 patients have been recruited, with 17 patients (17.9%) developing AL. Total cell count in drain fluid was significantly higher in AL patients on day 2, with a defined threshold yielding a negative predictive value (NPV) 0.88. There was no statistical difference between cell phenotypes. All cytokines (soluble biomarkers) measured showed significant differences between patients with and without AL on the first two post-operative days. Concentrations of IL-10 in drain fluid had the best diagnostic accuracy on day 1 (AUC = 0.75) with a defined low cut-off value yielding a NPV of 1 and a higher cut-off value yielding a positive predictive value of 0.67. In multivariable logistic regression, IL-10 remained significantly associated with AL, independently of age and sex. Drain fluid biomarkers such as total cell count and IL-10 show promise for AL prediction early post-operatively. Current qualitative work explores patient’s and healthcare professional’s views on barriers and facilitators to implementing drain fluid-based diagnostic tests.